3D Object detection is a fundamental task in vision-based autonomous driving. Deep learning perception models achieve an outstanding performance at the expense of continuously increasing resource needs and, as such, increasing training costs. As inference time is still a priority, developers usually adopt a training pipeline where they first start using a compact architecture that yields a good trade-off between accuracy and latency. This architecture is usually found either by searching manually or by using neural architecture search approaches. Then, train the model and use light optimization techniques such as quantization to boost the model’s performance. In contrast, in this paper, we advocate for starting on a much larger model and then applying aggressive optimization to adapt the model to the resource-constraints. Our results on large-scale settings for 3D object detection demonstrate the benefits of initially focusing on maximizing the model’s accuracy and then achieving the latency requirements using network pruning.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Hardware-Aware Latency Pruning for Real-Time 3D Object Detection


    Contributors:
    Shen, Maying (author) / Mao, Lei (author) / Chen, Joshua (author) / Hsu, Justin (author) / Sun, Xinglong (author) / Knieps, Oliver (author) / Maxim, Carmen (author) / Alvarez, Jose M. (author)


    Publication date :

    2023-06-04


    Size :

    19198314 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Real-time sleeper defect detection method based on pruning algorithm

    WU SONGRONG / TU ZHENWEI / YANG PING et al. | European Patent Office | 2022

    Free access


    PRUNING SYSTEM USING FLYING OBJECT

    TERASAWA MASAAKI / AKITA MINORU / KOYANAGI RANSHI et al. | European Patent Office | 2024

    Free access

    Rank-Based Filter Pruning for Real-Time UAV Tracking

    Wang, Xucheng / Zeng, Dan / Zhao, Qijun et al. | ArXiv | 2022

    Free access

    OPTIMAL SENSOR READING FOR LOW LATENCY OBJECT DETECTION

    HART SEAN | European Patent Office | 2022

    Free access